Acute lung injury, including the most severe form - acute respiratory distress syndrome (ARDS), is a devastating and, too often-times, lethal condition that can result from numerous seemingly unrelated direct or indirect pulmonary insults. Hundreds of proteins associated with the initiation or progression of the disease have been tested to delineate their possible role(s); however, little progress has been made to identify the key players responsible for its morbidity and mortality. Because mortality rates have changed little over the last decade using this candidate-gene approach, alternative strategies are essential to advance our understanding of the pathobiology of acute lung injury development and progression. The goal of this research is to use mouse models of acute lung injury to identify the major quantitative trait loci (QTLs) linked to mortality. To gain preliminary data for this proposal 18 common inbred mouse strains were screened for survival time in hyperoxia (>95% O2), a prototypic agent used to induce acute lung injury and ARDS. Two mouse models of differential survival time were identified. First, C57BL/6J (B) mice are sensitive, whereas 129X1/SvJ (S) mice are much more resistant to hyperoxia-induced acute lung injury mortality. Second, the resistant S strain was combined with 129P3/J (P), a closely related, but sensitive sub strain of the 129 line. Initial results with offspring generated for each mouse model suggested a complex mode of trait inheritance, including multiple genes and other genetic and epigenetic factor(s) (e.g., decreased penetrance, parental imprinting, and and/ or mitochondrial inheritance). S and P-derived crosses also suggested sex linkage. From these preliminary data, we hypothesize that hyperoxia-induced acute lung injury survival is a quantitative trait that is amenable to genetic analysis using inbred strains of mice to model the human disease. For each mouse model, the following three specific aims are proposed: (1) determine the likely mode of overall trait inheritance (segregation analysis) and estimate the number of loci contributing to the response; (2) identify genetic regions linked to hyperoxia-induced acute lung injury survival in backcross and F2 mice (QTL analysis) generated from strains of each model; and (3) identify candidate and positional candidate genes associated with the strain survival differences (microarray analysis). With this combined approach, we expect to gain insight into not only the pathology of hyperoxia-induced acute lung injury, but also the possible similarities to other oxidant-induced acute lung injuries. The proposed studies offer a different perspective to a retractable problem and could yield valuable information urgently needed to further assess genetic differences underlying disease risks and therapeutic outcomes in the population.